Editor’s Note:

Integrase strand transfer inhibitors (INSTIs) are associated with weight gain, but this weight gain doesn’t harm survival or increase cardiovascular disease (CVD) risk. So, does the weight gain associated with INSTIs lead to a higher risk of diabetes? At the recent 12th International AIDS Society (IAS) HIV Science Conference (IAS 2023), the renowned RESPOND cohort study from Europe and America indicated that INSTIs might increase the risk of diabetes onset. Even after adjustment, the onset risk remains statistically significant.

RESPOND Study: INSTI-related BMI Change and Diabetes Risk

People Living With HIV (PLWH) using INSTIs are related to an increase in Body Mass Index (BMI). A rise in BMI is also associated with a high risk of diabetes (DM). This study investigated the relationship between INSTI treatment, non-INSTI treatment, BMI changes, and diabetes risk.

Participants with CD4 cell counts, HIV RNA load, and multiple BMI measurements were included. Patients with a history of diabetes and pregnant women were excluded. Diabetes is defined as: random blood glucose >11.1 mmol/L, HbA1c >6.5%/48 mmol/mol, using anti-diabetic drugs or clinically diagnosed. A Poisson regression was used to assess the relationship between updated log BMI, current use of INSTI/non-INSTI, TDF/TAF, and their interactions with diabetes risk.

A total of 20,865 PLWH participated in the study, mostly males (74%) and Caucasians (73%). The median age was 45 (IQR: 37~52), and median BMI was 24 kg/m2 (IQR: 22~26). Out of 107,641 person-years of follow-up (PYFU), 785 new diabetes diagnoses occurred, with an incidence rate of 0.73/100 person-years (CI: 0.68~0.78). Log BMI was closely related to diabetes (aIRR 18.2, 95% CI: 11.7~28.3; P<0.001). In univariate analysis, current INSTI use was associated with an increased risk of diabetes (IRR 1.58, 95%CI: 1.37~1.82; P<0.001). When adjusted for updated log BMI and other variables, this association was somewhat reduced but remained statistically significant (aIRR 1.48, 95% CI: 1.28~1.72; P<0.001) (see figure below). In the adjusted analysis, current TAF and TDF users had similar diabetes risks (aIRR=0.98, 95%CI: 0.79~1.20, P=0.818). No evidence suggested an interaction between log BMI, INSTI, non-INSTI, and diabetes (P=0.130).

Among INSTI users, Raltegravir (RAL) accounted for 12%, Dolutegravir (DTG) for 60%, and other INSTIs for 28%, including Elvitegravir (EVG), Bictegravir (BIC), and Cabotegravir (CBG).

 The RESPOND study indicates that compared to non-INSTI users, INSTI users have an increased diabetes risk, which gets somewhat reduced after adjusting for BMI changes and other variables. There’s no difference in diabetes risk between TAF and TDF users.

Mixed Opinions: Does INSTI Increase Diabetes Risk?

It’s now relatively certain that INSTIs are linked to weight gain, but this gain doesn’t harm survival or increase CVD risk. The weight gain associated with INSTIs leading to an increased risk of diabetes is a matter of debate, with observational studies around the world showing varied results.

Insulin resistance (IR) is a precursor of diabetes, and current studies on the association between INSTI and IR are inconsistent. For instance, the AIDS Clinical Trials Group’s A5260s study analyzed 328 patients and found that patients starting on TDF/FTC treatment had a rapid increase in IR that then stabilized; compared to ATV/r or DRV/r, RAL showed no difference in terms of IR. The TANGO study showed a positive effect with DTG/3TC.

Regarding diabetes, a recent observational cohort study (SCOLTA) published in the “AIDS” journal, analyzed 4,366 patients. During the follow-up (January 2003 to November 2021), 120 new diabetes cases were diagnosed, with an incidence rate of 1.26 cases/100 person-years. This research found that baseline weight was significantly associated with diabetes onset (annual risk aHR 1.03; 95%CI: 1.01~1.06), not the weight gain during INSTI treatment. Other factors, such as older age (aHR 1.03), untreated hypertension (aHR 2.90), and baseline fasting blood glucose >100 mg/dl (aHR 5.47), all increased diabetes risk, consistent with the general population’s risk factors. Thus, this study concluded that INSTIs aren’t related to an increased diabetes risk.

A systematic review and meta-analysis published in BMJ Open Diabetes Res Care included 13 studies with 72,000 patients, showing that new-onset diabetes risk in INSTI-treated patients reduced by 20% (HR 0.80, 95%CI: 0.67~0.96); compared to Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTIs), the risk reduced by 25% (HR 0.75) and reduced by 22% compared to Protease Inhibitors (PIs).